20110621

Agenda

  • Review ACTION points from previous meeting
    • [ACTION] sommer and j1mc to look at doc presentation
  • Oneiric Development
  • Ubuntu Server Team Events
  • Weekly Updates & Questions for the QA Team (hggdh)

  • Weekly Updates & Questions for the Kernel Team (smb)

  • Weekly Updates & Questions regarding Ubuntu ARM Server (NCommander)

  • Weekly Updates & Questions from the Ubuntu Community

  • Open Discussion
  • Announce next meeting date and time

Minutes

Meeting Actions

  • sommer and j1mc to look at doc presentation - carried forward

Review ACTION points from previous meeting

  • sommer and j1mc to look at doc presentation - carried forward

Oneiric Development

See http://status.ubuntu.com/ubuntu-oneiric/group/topic-oneiric-server-overview.html for detail.

  • Merges and syncs going great
  • New bugs manageable
  • Need to focus on new feature delivery
    • MIR's for OpenStack - loads of them; all team to help out where possible - ideally all MIR's should be raised by the end of the week.

Ubuntu Server Team Events

  • zul - Linuxcon Augst
  • SpamapS - OSCON, Portland next month

Weekly Updates & Questions for the QA Team (hggdh)

  • SRU verification for TGT and NBD
  • Automated testing toolsets will need updating for A2 to accommodate installer changes
  • Ubuntu QA Lab setup continues.

Weekly Updates & Questions for the Kernel Team (smb)

  • Xen support - query around which kernel should be uses; must support blkback and netback; will be added to init.
  • Hardy AMI (https://launchpad.net/bugs/705562) - solution identified - see bug report for details.

Weekly Updates & Questions regarding Ubuntu ARM Server (NCommander)

  • NCommander not present

Weekly Updates & Questions from the Ubuntu Community

  • No updates or questions

Open Discussion

  • SpamapS: call for more ensemble formulas to be written.
  • adam_g: OpenStack in Oneiric needs more testing - find time to try it out.

Agree on next meeting date and time

  • Next meeting will be on Tuesday, June 28th at 16:00 UTC in #ubuntu-meeting.

Log

[17:02] <jamespage> #startmeeting
[17:02] <MootBot> Meeting started at 11:02. The chair is jamespage.
[17:02] <MootBot> Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE]
[17:02] <jamespage> [TOPIC] Review ACTION points from previous meeting
[17:02] <MootBot> New Topic:  Review ACTION points from previous meeting
[17:02] <jamespage> sommer and j1mc to look at doc presentation
[17:02] <jamespage> sommer or j1mc around?
[17:03] <jamespage> hmm guess not
[17:03] <Daviey> jamespage: sorry, i am here.
[17:03] <jamespage> [TOPIC] Oneiric Development
[17:03] <MootBot> New Topic:  Oneiric Development
[17:03] <Daviey> Hello!
[17:03] <jamespage> just in time - all yours
[17:03] <Daviey> http://status.ubuntu.com/ubuntu-oneiric/group/topic-oneiric-server-overview.html
[17:03] <adam_g> o/
[17:03] <MootBot> LINK received:  http://status.ubuntu.com/ubuntu-oneiric/group/topic-oneiric-server-overview.html
[17:03] <Daviey> (/me actions himself to get the trend line reset)
[17:03] <Daviey> Firsty, we are doing much better with our merging and syncing; and our new bug count is looking better.  However, whilst we are making progress burning work items; we possibly need to switch more focus to the feature development.
[17:03] <Daviey> Incidently, we have a metric tonne of MIR's we need to undertake.  We really need to distribute these agmonst the team.  We should strive to raise the MIR's for at least two each before next week.
[17:04] <Daviey> Takeway.. if there isn't a WI for the engineering effort currently being done; ask yourself if you should be doing it. :)
[17:04] <Daviey> ...
[17:05] <jamespage> Daviey: are most of the MIR's around the dependencies for openstack?
[17:05] <Daviey> jamespage: mostly yes
[17:05] <Daviey> but TBH, there shouldn't be too much of a requirement to have a deepo knowledge of the package
[17:05] <jamespage> rightoh
[17:06] <Daviey> so pick 2.. and just get 'em done.
[17:06] <Daviey> Remember to mark INPROGRESS
[17:06] <Daviey> :)
[17:06] <zul> oooh...i can stop then :)
[17:06] <Daviey> (sorry for brevity, on the phone at the moment)
[17:06] <Daviey> zul: no, you need to do 4.
[17:06] <zul> oooh...i can stop then :)
[17:06] <jamespage> any other questions for Daviey or about oneiric development?
[17:06] <SpamapS> Daviey: briefly, I think the new bug count is actualy getting worse, not better.
[17:06] <Daviey> SpamapS: it's better than Friday.
[17:07] <Daviey> :)
[17:07] <Daviey> SpamapS: But yes, it's something to watch.. However, at the moment, we really do need to put more focus into feature work.
[17:07] <SpamapS> Daviey: 42 and 62 aren't really that much different. We were at 0 for 2 entire cycles.
[17:08] <zul> SpamapS: i was cracking at them this morning
[17:08] <hallyn> so the q is, is everyond doing their half-day bug triage, and is that not being enough?
[17:08] <SpamapS> I know that we have had a few more packages added, but we haven't had resources added.
[17:08] <Daviey> SpamapS: Hmm.. I have a graph.. but yes, but count is raising. And there is a secret weapon to help address this.
[17:08] <Daviey> hallyn: I think we've all fallen behind on the half day triage.
[17:09] <hallyn> Daviey: well that's good news bc it means maybe if we are more disciplined abouthat we can address this without being more drastic :)
[17:09] <smoser> [late question: where is the MIR list]
[17:09] <zul> hallyn: im not i know i been slacking
[17:09] <Daviey> smoser: see the WI's
[17:09] <smoser> ah
[17:09] <Daviey> smoser: http://status.ubuntu.com/ubuntu-oneiric/group/topic-oneiric-server-overview.html (search for "MIR ")
[17:10] <jamespage> OK sounds like we all need to ensure that we are doing the full half day of bug triage that we sign up for each week
[17:10] <jamespage> it would be good to get the new bug count down
[17:10] <hallyn> Mind you I think the wiki page suggests it shoud only take a few minutes  :)
[17:10]  * smoser is looking forward to a bug triage engineer
[17:10] <smoser> there are major trends in that bug count
[17:11] <hallyn> smoser: are you cloning one? (an engineer that is)
[17:11] <smoser> that surely can be identified and easily reduced / removed with some dedicated engineering
[17:11] <smoser> hallyn, i thought i had put you in charge of that.
[17:11] <jamespage> anything else on Oneiric development?
[17:11] <Daviey> jamespage: no, please move on.
[17:11] <SpamapS> I also think that we may be writing off some common recurring themes as "a system problem" when the pattern itself suggests we should eliminate the automated bug report.
[17:11] <hallyn> smoser: mine will take, oh, 18 years to be ready.
[17:11] <jamespage> [TOPIC] Ubuntu Server Team Events
[17:11] <hallyn> but she's working on it
[17:11] <MootBot> New Topic:  Ubuntu Server Team Events
[17:11] <jamespage> anyone attending anything interesting?
[17:11] <hallyn> SpamapS: good point, maybe we should talk about that sometime
[17:12] <zul> linuxcon in august
[17:12] <SpamapS> yeah thats a topic for next week.
[17:12] <hallyn> ok
[17:12] <smoser> http://ubuntu-server-new-bugs.notlong.com/
[17:12] <MootBot> LINK received:  http://ubuntu-server-new-bugs.notlong.com/
[17:12] <SpamapS> Probably worth talking about now that its about a month away.. I'll be a OSCON in Portland
[17:12] <smoser> (juust for the record)
[17:13] <jamespage> noted
[17:13] <jamespage> [TOPIC] Weekly Updates & Questions for the QA Team (hggdh)
[17:13] <MootBot> New Topic:  Weekly Updates & Questions for the QA Team (hggdh)
[17:13] <jamespage> hggdh: around?
[17:13] <hggdh> yes
[17:13] <jamespage> all yours
[17:15] <hggdh> no news from me, except a SRU verification fot TGT and NBD
[17:15] <hggdh> ..
[17:15] <Daviey> hggdh: Thanks for that.
[17:15] <Daviey> hggdh: We should probably sync up soon to look at what needs testing for A2.
[17:16] <Daviey> A2 should have installer differences.
[17:16] <hggdh> Daviey: indeed. We will need to revisit all tests
[17:16] <Daviey> hggdh: oh joy. :)
[17:16] <jamespage> hggdh: and update the automated testing as required as well
[17:16] <Daviey> hggdh: How is jenkins deployment going for CI testing?
[17:17] <hggdh> we are moving all Jenkins to the new lab facility
[17:17] <hggdh> and will be (hopefully) expanding the tests
[17:18] <jamespage> Daviey: when are the installer changes due to land?
[17:19] <Daviey> jamespage: by end of next week
[17:19] <jamespage> right so we need to line up any automated test changes to that timescale otherwise it will all break....
[17:20] <jamespage> ok anything else for hggdh?
[17:20] <jamespage> [TOPIC] Weekly Updates & Questions for the Kernel Team (smb)
[17:20] <MootBot> New Topic:  Weekly Updates & Questions for the Kernel Team (smb)
[17:20] <smb> Hi
[17:20]  * SpamapS makes a note not to write his boot testing until after the installer changes
[17:20] <smb> So a bit more playing with Oneiric dom0. There is one thing I think we need to decide.
[17:22] <smb> Last time I looked, the xen tools would use the -server kernel installed as the dom0 kernel. Should that be changed to -virtual? With the benefit that -virtual should probably have blkback and netback compiled in. Which is currently a pita when trying to launch domUs
[17:23] <SpamapS> err
[17:23] <SpamapS> dom0 is a real server, so -virtual wouldn't have the needed components
[17:23] <zul> which xen-tools?
[17:24]  * SpamapS reiterates his long time opinion that this is why kvm > xen :-P
[17:24] <smoser> dom0 is not a real server
[17:24] <smb> Hm, true. (somehow was looking at hardy which take a xen kernel for both, but probably not a good example then)
[17:24] <zul> SpamapS: centos fan boy ;)
[17:24] <smoser> but i think -virtual is probably not sufficient
[17:24] <zul> right it isnt
[17:25] <smoser> do you happen to know  why?
[17:25] <zul> is that directed at me?
[17:25] <SpamapS> smoser: huh?
[17:26] <smoser> dom0 is not a "real server".
[17:26] <smoser> xen runs on the "real server"
[17:26] <SpamapS> Dom0, or domain zero to expand the abbreviation, is the first domain started by the Xen hypervisor on boot. It has special privileges, like being able to cause new domains to start, and being able to access the hardware directly. Unless DriverDomains are being used, it is responsible for running all of the device drivers for the hardware.
[17:26] <smoser> dom0 runs under xen
[17:26] <SpamapS> http://wiki.xensource.com/xenwiki/Dom0
[17:26] <MootBot> LINK received:  http://wiki.xensource.com/xenwiki/Dom0
[17:26] <SpamapS> smoser: effectively, its a real server
[17:26] <smoser> it is meerley a priviledged domU
[17:26] <SpamapS> it needs all the drivers and hardware bits to be useful in the general sense.
[17:27] <SpamapS> Tho I agree it is not *actually* the real kernel.
[17:27] <Daviey> Is this OT?
[17:27] <smoser> yes.
[17:27] <smoser> i was mostly wrong.
[17:27] <smb> Ok accepted- So the question would rather be to make sure blkback and netback are really loaded when xen loads that kernel. (maybe in init?)
[17:27] <smoser> i was kind of thinking that dom0 might nit need drivers in some cases.
[17:27] <smoser> but was wrong
[17:28] <smb> Not sure we want that built-in into the kernel.
[17:28] <zul> smb: can you open up a bug against xen in launchpad about it then i can add it init
[17:28] <smb> zul, ok can do that
[17:28] <zul> thanks
[17:28] <smb> So the other thing I wasted most of my last week is bug 705562
[17:28] <ubottu> Launchpad bug 705562 in linux (Ubuntu Hardy) "ami-6836dc01 8.04 32 bit AMI kernel lock bug" [Undecided,In progress] https://launchpad.net/bugs/705562
[17:29] <smb> And finally I think I got a solution
[17:29] <smb> ... and its a bit scary
[17:30] <smb> It made me wonder whether 32bit was ever really used. Anyway, thats all.
[17:30] <jamespage> anyone have any questions for smb?
[17:31] <Daviey> smb: On another note, did you have thoughts on what we discussed yesterday; regarding the reserved memory theft?
[17:31] <Daviey> How we can add optimisations for low memory systems?
[17:31] <smb> Daviey, well not more than I was saying. Givem more. That actually is what older versions really did
[17:32] <smb> There is a price to pay to have pae and a virtual memory management
[17:33] <Daviey> smb: Okay, i'm interested if we can make cgroups disable memory hogging more discovrable for small systems
[17:33] <smb> Daviey, and we need to be careful about what this does. The hint is given in dmesg
[17:33] <smoser> smb, 32 bits are used.
[17:33] <smoser> definitely
[17:33] <Daviey> smb: okay, thanks.
[17:33] <smoser> at least on amazon, price makes them a real usable option
[17:34] <smb> smoser, Right, well the comment was about hardy version
[17:34] <Daviey> smb: that is all from me, thanks!
[17:34] <SpamapS> m1.small is 32-bits, yes?
[17:34] <smb> And about the fact that what happens breaks things quite badly and wondering how it worked at all (for longer period of time)
[17:35] <smb> SpamapS, right it is
[17:35] <SpamapS> Quite a popular instance size :-P
=== Quintasan_ is now known as Quintasan
[17:36] <jamespage> OK; moving on
[17:36] <jamespage> [TOPIC] Weekly Updates & Questions regarding Ubuntu ARM Server (NCommander)
[17:36] <MootBot> New Topic:  Weekly Updates & Questions regarding Ubuntu ARM Server (NCommander)
[17:36] <Daviey> NCommander: are you here?
[17:36] <jamespage> NCommander: around?
[17:37] <jamespage> hmm I would guess not
[17:37] <jamespage> anyone else able to give and update on ARM server?
[17:38]  * SpamapS hopes
[17:38] <jamespage> OK guess not
[17:38] <jamespage> moving on
[17:39] <jamespage> [TOPIC] Weekly Updates & Questions from the Ubuntu Community
[17:39] <MootBot> New Topic:  Weekly Updates & Questions from the Ubuntu Community
[17:39] <jamespage> So any questions or updates from anyone in the ubuntu-server community?
[17:40] <jamespage> Nope
[17:40] <jamespage> [TOPIC] Open Discussion
[17:40] <MootBot> New Topic:  Open Discussion
[17:41] <jamespage> anyone have anything else they want to discuss/raise/inform us all about?
[17:41] <SpamapS> Hey everybody write more ensemble formulas! :)
[17:41] <RoAkSoAx> JAMES PAGE!
[17:41] <SpamapS> ETHERPAD
[17:42] <zul> jaaaaaaaaaaameeeeeeeeessssssssss page!
[17:43] <Daviey> thanks jamespage
[17:43] <robbiew> JAMES PAGE!!!1
[17:43] <adam_g> openstack and its related components can use more testing in oneiric! since ive been using it ive hit two major bugs that could have been discovered by simply trying to run it
[17:43] <zul> that glance bug is fixed btw
[17:43] <Daviey> adam_g: good point
[17:43] <adam_g> zul, i saw. thanks!
[17:43] <smoser> JAMES PAGE is an ubuntu developer. Woot!
[17:43] <robbiew> +10 adam_g
[17:44] <robbiew> I guess his developer contributions and activity were in line with precedence
[17:44] <robbiew> lol
[17:44] <RoAkSoAx> lol
=== Guest72525 is now known as Omega`
[17:44] <jamespage> [TOPIC] Announce next meeting date and time
[17:44] <MootBot> New Topic:  Announce next meeting date and time
[17:44] <robbiew> (only funny if you were in yesterday's ubuntu membership meeting)
[17:45] <jamespage> Tuesday, June 28th 2011 16:00 UTC
[17:45] <jamespage> see you then
[17:45] <jamespage> #endmeeting

MeetingLogs/Server/20110621 (last edited 2011-06-23 14:30:50 by james-page)